Modern data systems typically include many types of computers. In order to maximize use of physical computing resources, many of these data systems include virtualized machines. A single machine may potentially be turned into, for example, 10 virtual machines.
To ensure no loss of data on each of the virtual machines, each of the virtual machines will tend to require its own backup on a backup server. Users requiring access to these backups may initiate multiple requests to connect to the same backup.
By initiating multiple connections to a particular backup, users may potentially create inconsistent operational requests. For example, a user may attempt to restore the contents on backup “B16” to virtual machine “V16.” The restore process might be lengthy and that same user may attempt to initiate a new connection an hour later to the same backup “B16” in order to modify or delete a single file. Because this additional request has been made during the processing of the restore request, one or both of these request operations may now fail.
Also, users may leave one or more connections to the backup open for an indeterminate amount of time. The process used to display the file system structure present on the backup may be a resource intensive process and can negatively impact performance of the operating system if it runs longer than necessary.
There is a need, therefore, for an improved method, article of manufacture, and apparatus for managing stateless processes.